Deep ensemble multitask classification of emergency medical call incidents combining multimodal data improves emergency medical dispatch
作者：
Highlights：
• Deep learning outperforms clinical rules in emergency medical calls classification.
• There is unused latent predictive information in emergency medical calls registries.
• Free text dispatcher observations contain information of high predictive value.
• Machine learning based decision support can improve emergency medical dispatch.
摘要
•Deep learning outperforms clinical rules in emergency medical calls classification.•There is unused latent predictive information in emergency medical calls registries.•Free text dispatcher observations contain information of high predictive value.•Machine learning based decision support can improve emergency medical dispatch.
论文关键词：Medical emergencies,Emergency medical calls,Emergency medical dispatch,Deep learning,Ensemble learning,Multitask learning
论文评审过程：Received 9 November 2020, Revised 19 April 2021, Accepted 3 May 2021, Available online 13 May 2021, Version of Record 21 May 2021.
论文官网地址：https://doi.org/10.1016/j.artmed.2021.102088
